mi_analyze_document: Run a one or more documents through a misinformation/bias...

In hrbrmstr/misinfo: Tools to Perform 'Misinformation' Analysis on a Text Corpus

Description

Run a one or more documents through a misinformation/bias sentiment analysis

Usage

mi_analyze_document(path = ".", pattern = NULL, bias_type, sentiment_list)

Arguments

path	a character vector of full path names; the default corresponds to the working directory, getwd(). Tilde expansion (see 'path.expand“) is performed. Missing values will be ignored.
pattern	an optional regular expression. Only file names which match the regular expression will be returned. (e.g. ".*\.txt$)
bias_type	a one-word (no punctuation or spaces) description of the bias type of the sentiment list being used. This will be used to tag the output so it can be used later when aggregating results of multiple bias-type analyses.
sentiment_list	a character vector of words/phrases. The built-in sentiment bias lists can be used via mi_use_builtin() or the caller can pass in a charcter vector of lower-case words/phrases

Value

a list (classed as misinfo_corpus) with:

• bias_type (the bias type label passed in to the function)

• corpus (the raw data from the processed document corpus)

• frequency_summary (frequncy summary for each document in corpus)

• individual_frequency (individual component frequency for each phrase in each document in the corpus)

Examples

mi_analyze_document(
  system.file("extdat", package="misinfo"),
  pattern=".*txt$",
  bias_type = "explanatory",
  sentiment_list = mi_use_builtin("explanatory")
)

hrbrmstr/misinfo documentation built on May 7, 2019, 9:36 a.m.